Malignant glioma, the most common primary malignant brain tumor in adults, is debilitating and rapidly fatal. Very little is known about its causes; family history and diverse environmental factors have been suggested in epidemiologic studies and several genes have been implicated in molecular biology studies. We are currently conducting a large, NCI-funded, population-based epidemiologic study of malignant glioma in the San Francisco Bay Area to examine and compare family and personal medical histories, occupation, smoking, alcohol use, diet, electromagnetic field exposures and other characteristics in cases and controls. We are conducting detailed interviews with cases and age-, sex-, and race-matched controls. However, there is currently no funding for blood collection or for genotype determination for pertinent oncogenes. A major purpose of this small grant is to collect, process, and store blood samples from 300 malignant glioma cases as we accrue them and 300 controls to enable molecular genetic studies of these subjects as candidate genes become identified. Because the most common form of malignant glioma is rapidly fatal, we would be unable to draw blood from these cases in the future; thus, there is time-limited access to cases in this study. Additional aims of this small grant are to: determine Ha-ras genotypes of cases and controls; test whether rare Ha-ras alleles are more common in cases, as they were in one previously published study; and use the information collected on Ha-ras to serve as a model for developing a comprehensive proposal for the evaluation of candidate genes. An experienced phlebotomist will collect blood from willing subjects. Standard methods will be used to prepare genomic DNA samples from white blood cells and to conduct Southern blot studies to determine the Ha-ras genotypes in these samples. We will have adequate power to detect meaningful differences in the proportion of rare Ha-ras VTR alleles in cases and controls. In addition to the genetic data, information collected on other characteristics of subjects will be used to explore whether any of these factors modify the association of Ha-ras VTR rare alleles with malignant glioma. This study will be the first to combine genotyping of individuals with and without malignant glioma with regard to one suspected susceptibility gene in an epidemiologically well-defined population-based study. It will therefore serve as a pilot study for a more comprehensive evaluation of other susceptibility genes for malignant glioma.